The recently developed natural radon tracer method has potential as a rapid, low‐cost, nondestructive, and noninvasive method for quantifying NAPL contamination. In the subsurface, radon‐222 (radon) is produced by the decay of naturally occurring radium‐226 contained in the mineral fraction of aquifer solids. In groundwater radon occurs as a dissolved gas, with a half‐life of 3.83 days. In the absence of NAPL, the radon concentration in groundwater quickly reaches a maximum value that is determined by the mineral composition of the aquifer solids, which controls the rate of radon emanation. In the presence of NAPL, however, the radon concentration in the groundwater is substantially reduced due to the preferential partitioning of radon into the organic NAPL phase. A simple equilibrium model and supporting laboratory studies show the reduction in radon concentration can be quantitatively correlated with residual NAPL saturation. Thus, by measuring the spatial distribution in radon it may be possible to identify locations where residual NAPL is present and to quantify the NAPL saturation. When the basic processes of partitioning, radon emanation from the aquifer solids, and first‐order decay are incorporated into an advective/dispersive transport model, good agreement is obtained with the results of laboratory and field experiments. Model sensitivity analyses shows many factors can contribute to the radon concentration response, including the length of the NAPL zone, NAPL saturation, groundwater velocity, porosity, and radon emanation. Thus, care must be taken when applying the radon method to locate and quantify NAPL contamination in the subsurface.